Fixes for the following issues will be released shortly:
CUDNN Flash attention disabled by default until further testing is done
For data preparation of GPT models, use your own dataset or an online dataset legally approved by your organization
Megatron Core gradient clipping issue when using MoE and EP
Race condition in NeMo experiment manager
Mistral & Mixtral tokenizers require Hugging Face login
Gemma, Starcoder, and Falcon 7B export to TRT-LLM only works with a single GPU and if the user wants to export, there is no descriptive error message shown to the user.
The following notebooks have functional issues and will be fixed in the next release
ASR_with_NeMo.ipynb
ASR_with_Subword_Tokenization.ipynb
AudioTranslationSample.ipynb
Megatron_Synthetic_Tabular_Data_Generation.ipynb
SpellMapper_English_ASR_Customization.ipynb
FastPitch_ChineseTTS_Training.ipynb
NeVA Tutorial.ipynb